DSXchange

the line numbers are rarely correct, they are usually off by a couple of lines. If you post lines 325 through 340 or look into those lines you will find the appropriate error.
Does the job abort immediately or does it take a while before it aborts?

That should be quite easy to do. Instead of me trying to figure out the c++ code for the decimal unpacking, could you explain it to me in words?

No, not that I know of.

If you have used the table metadata everywhere, then you can do a find from within DataStage, otherwise it is more difficult. You could export the project and search for the table name in the .dsx or .xml file.

Back again - if you don't change the default, but set a specific format for your date column, does it work?

Vijay, at least once a week someone posts pretty much exactly the same question on DSXChange. Please use the Search function before posting a question, particularly if it is not a very exotic type one; DSXChange has been around for many years and there is a lot of stored knowledge in the old threads...

It might be best to store this last-run-date in your database as opposed to a sequential file, that depends on whether or not other processes need to know the last run date.

Ignore the errors after the first one, they are just window dressing. Your first error was ...Occurred: 1:13:42 AM On date: 11/15/2009 Type: Warning Event: SRC_EXT_STG_QUOTA_PULLBACKS,0: *** glibc detected *** /opt/IBM/InformationServer/Server/PXEngine/bin/osh: malloc(): memory corruption (fast): 0x...

If I understand your post correctly, you changed the job default setting for date format in one of the job tabs, right? I have had trouble with this in past as well and ended up using a specific column format. This problem was at an older V7 version and might have since been fixed.

The "LOAD" method requires exclusive access, unlike the "upsert" mode, talk to your DBA to check on who is holding locks to the table when you try a bulk load.

Your best best is to get a DB2 dba. Then see if you can successfully do a "db2 connect" from your UNIX machine (DataStage server); once that functions you know that, when correctly configured, DataStage will also be able to.

"df ." when you were attached to which directory? Can you run your job and monitor the usage in your scratch directory?

An urgent problem... and only for Ray... This could be interesting.

"OE" is the Oracle Enterprise Stage.

If you put an echo in the bat file does that print out to the log, i.e. does the BAT file get called correctly?